sensors

Article

Studies of NO2 Gas-Sensing Characteristics of aNovel Room-Temperature Surface-Photovoltage GasSensor Device

Monika Kwoka * and Jacek Szuber

Silesian University of Technology, Faculty of Automatic Control, Electronics and Computer Science,Department of Cybernetics, Nanotechnology and Data Processing, 44-100 Gliwice, Poland; Jacek.Szuber@polsl.pl* Correspondence: monika.kwoka@polsl.pl; Tel.: +48-32-237-20-57

Received: 1 December 2019; Accepted: 6 January 2020; Published: 11 January 2020�����������������

Abstract: In this work the characteristics of a novel type of room temperature NO2 gas sensor devicebased on the surface photovoltage effect are described. It was shown that for our SPV gas sensordevice, using porous sputtered ZnO nanostructured thin films as the active gas sensing electrodematerial, the basic gas sensor characteristics in a toxic NO2 gas atmosphere are strongly dependent onthe target NO2 gas flow rate. Moreover, it was also confirmed that our SPV gas sensor device is ableto detect the lowest NO2 relative concentration at the level of 125 ppb, with respect to the commonlyassumed signal-to-noise (S/N) ratio, as for the commercial devices.

Keywords: surface photovoltage gas sensor; room temperature working conditions; gas sensorcharacteristics; NO2 atmosphere

1. Introduction

After more than six decades of development only a few types of gas sensors devices have beenconstructed. These are based on a variety of chemical and physical methods including electrochemical,catalytic, optical, acoustical and electrical designs. Reviews of the most common types of gas sensorsinclude, among others, those of Comini et al [1] and by Xiu et al [2].

Among the electrical methods, the most common conductometric type gas sensor devices(systems) are based on semiconductor metal oxides (MOX), organic materials, carbon nanotubes andconductometric polymers. However, even after five decades of development, they still exhibit somecritical and fundamental limitations, such as high-temperature working conditions and related highpower consumption. These sensors do have good sensitivity; depending on the gas, usually at thelevel of ppm, but this is combined with a rather poor selectivity (a trait that can be slightly improvedupon by adding noble catalytic metals); in addition, they possess rather poor dynamic parameters(long response and recovery times) [3–12].

In order to overcome the above-mentioned crucial limitations of conductometric MOX gas sensors,some innovative ideas and related approaches have appeared in the literature in recent years describingimprovements of their sensing abilities.

One of these methods exploits the effect-of-work function (contact potential difference (CPD))variation, which can be measured by using the Kelvin vibrating capacitor as a transducer [13,14].However, because of rather poor gas sensitivity due to a rather low signal-to-noise (S/N) ratio, thismethod has received only limited application in studies of the porous semiconductor materials asrecently reviewed by Korotcenkov [15].

This type of sensor can be improved by using the additional external illumination of the gassensor material to play the role of measuring electrode, a phenomenon commonly known the surface

Sensors 2020, 20, 408; doi:10.3390/s20020408 www.mdpi.com/journal/sensors

Sensors 2020, 20, 408 2 of 9

photovoltage (SPV) effect [16]. However, until now the SPV effect has received surprisingly limitedapplication for the detection of specific gases at the surfaces of selected metal oxide materials [17,18].

In our recent paper [19], we have briefly described the fundamentals of SPV effect and its potentialapplication for gas sensing, together with a proof-of-principle of our novel type SPV gas sensor systembased on the Kelvin probe approach using porous ZnO nanostructured thin films as an active electrode,and a flat Cu metallic grid-type as the reference electrode. Moreover, the basic analytical abilities of ourSPV gas sensor device for the detection of nitrogen dioxide NO2 atmosphere in relative concentrationsof up to 1 ppm in synthetic air have been verified and interpreted on the basis of information on localsurface chemistry and morphology of porous ZnO nanostructured thin films.

In this paper the results of systematic studies of the usable gas-sensing characteristics of ournew SPV gas sensor device are presented, including the determination of the influence of gas flowrate of NO2 at constant relative concentration in synthetic air on the variation of SPV amplitudeversus time, as well as the respective gas sensor dynamic parameters, such as response and recoverytimes. The influence of gas flow rate at constant relative concentration on the gas sensor responsecharacteristics has recently been investigated by several groups, among others by Eklöv et al. [20]for the Pd-MOSFET sensors in H2 detection, by Lezzi et al. [21] for the RGTO SnO2 film-basedconductometric sensor for CO detection, by Utriainen et al. [22] in the comparative analysis of aminiaturized ion mobility spectrometer and metal oxide gas sensor for the detection of toxic organicvapors, by Kevin et al. [23] for SnO2 film-based conductometric sensor mainly for CO detection, byGmür et al. [24] for the metal-oxide-based gas sensor microarrays mainly for isopropanol detection,and by Righettoni et al. [25] for portable WO3 gas sensors for breath analysis.

Moreover, we have determined the variation in amplitude of the SPV signal with decreasingrelative concentrations of NO2 in synthetic air well below 1 ppm level at the chosen middle constantgas flow rate, to attain the lowest NO2 relative concentration that can be experimentally detected at thecommonly assumed signal-to-noise (N/S) ratio of 3.

2. Materials and Methods

In our experiment the recently elaborated SPV gas sensor device [19] was used based on theporous ZnO-nanostructured thin films [26,27] as an active electrode, and the flat Cu metallic grid-typereference electrode mounted inside the rectangular SPV gas sensor measuring chamber of dimensions20 × 40 × 30 mm, which were matched to the dimensions of the Kelvin probe flat type vibratingcapacitor system used. The SPV signal response after UV illumination of the ZnO active electrode bythe UV5-400-30 type LED diode was measured by the microcontroller data processing and respectiveacquisition system [19].

For the determination of the lowest NO2 relative concentration in dry synthetic air significantlybelow the ppm level we used a novel gas dilution, mixing and dosing system, that was designed byour group and constructed by the MEDSON Company (Paczkowo, Poland). Its simplified schematicidea is shown in Figure 1.

In general, this system consists of:

- two bottles and related stainless steel channels of NO2 toxic gas at two different startingconcentrations in synthetic air (50 ppm to reach a relative concentration range above 1 ppm forprimary measurements, and 1 ppm to reach a relative concentration range below 1 ppm for finestmeasurements, respectively), combined with a set of respective cut-off valves;

- the gas dilution, mixing and dosing parts based on gas mass flow controllers (MFC)–model BrooksSLA5850 (Brooks Instruments, Hatfek, PA, USA) combined with a set of respective cut-off valves;

- gas mixing chamber with a set of baffle peers to reach the well defined precise final constantrelative concentration of NO2 gas mixture in the synthetic air;

- “back pressure” unit based on the gas mass flow controllers (MFC)–model Brooks SLA5820(Brooks Instrument, Hatfek, PA, USA) as the pressure regulators, combined with two exhaust

Sensors 2020, 20, 408 3 of 9

systems equipped with additional toxic gas washing bottles (to avoid any undesired escape ofNO2 toxic gas mixture to the surrounding atmosphere).

In relation to the above, it should be added that this system is also equipped with a microcontrollerprocessing unit for its control, combined with the data processing and acquisition, working with therespective software (MEDSON FC).

Sensors 2020, x  3  of  9 

In relation to the above, it should be added that this system is also equipped with a microcontroller 

processing unit for its control, combined with the data processing and acquisition, working with the 

respective software (MEDSON FC). 

 

Figure 1. Simplified block‐scheme of NO2 gas dilution, mixing and dosing system for use with the 

surface photovoltage (SPV) gas sensor system (the device). 

It should be underlined that in our experiments we used a gas handling procedure allowing the 

repeatable measuring conditions defined by the specific NO2 gas flow rate in combination with the 

specific NO2 relative concentration in synthetic air reaching the SPV gas sensor measuring chamber. 

It consists of two independent steps, i.e. the controllable value of the NO2 gas flow rate which was 

achieved by using the above described gas dilution, mixing and dosing system (MEDSON), whereas 

the controllable NO2 relative concentration was achieved by using additionally the GasAlert Extreme 

NO2 detector (Honeywell BW Technologies, Calgary, AB, Canada). For the evaluation of relative NO2 

concentration below 1 ppm (its detection threshold) an extrapolation procedure was used. Finally, 

between  the  subsequent gas  sensing measurements  the SPV gas  sensor measuring  chamber was 

rinsed with pure synthetic air for 30 min.   

3. Results and Discussion 

In our research, we have  focused on  the working conditions of our highly sensitive SPV gas 

sensor prototype, which can be divided in the following two steps: 

‐ determination of the influence of NO2 gas flow rate at constant relative concentration on the gas 

sensor response characteristics, together with basic gas sensor dynamic parameters; 

‐ determination of the lowest NO2 relative concentration that can be detected by our SPV gas sensor 

device with respect to the commonly used signal‐to‐noise (S/N) ratio, one of the most important 

usable parameters for most commercial gas detection systems. 

3.1. Influence of NO2 Gas Flow Rate at Constant Relative Concentration on Gas Sensor Response   

In general,  taking  into account  the dynamic aspects of gas adsorption/desorption effects at  the 

surface of gas sensor materials, the gas‐sensing characteristics should strongly depend on the flow rate 

of target gas reaching the gas sensor system measuring chamber. This is related to the fact that with an 

increased flow rate, the amount of NO2 adsorbed at the surface of the gas sensor material also increases. 

This  can be a  cause of  the variation of  shape of  response  curve  towards  the  corresponding higher 

response time, as well as the shape of recovery curve towards the corresponding higher recovery time 

directly related to the favorable conditions for the gas desorption. Thus, in our study we have determined 

Figure 1. Simplified block-scheme of NO2 gas dilution, mixing and dosing system for use with thesurface photovoltage (SPV) gas sensor system (the device).

It should be underlined that in our experiments we used a gas handling procedure allowing therepeatable measuring conditions defined by the specific NO2 gas flow rate in combination with thespecific NO2 relative concentration in synthetic air reaching the SPV gas sensor measuring chamber.It consists of two independent steps, i.e. the controllable value of the NO2 gas flow rate which wasachieved by using the above described gas dilution, mixing and dosing system (MEDSON), whereasthe controllable NO2 relative concentration was achieved by using additionally the GasAlert ExtremeNO2 detector (Honeywell BW Technologies, Calgary, AB, Canada). For the evaluation of relative NO2

concentration below 1 ppm (its detection threshold) an extrapolation procedure was used. Finally,between the subsequent gas sensing measurements the SPV gas sensor measuring chamber was rinsedwith pure synthetic air for 30 min.

3. Results and Discussion

In our research, we have focused on the working conditions of our highly sensitive SPV gas sensorprototype, which can be divided in the following two steps:

- determination of the influence of NO2 gas flow rate at constant relative concentration on the gassensor response characteristics, together with basic gas sensor dynamic parameters;

- determination of the lowest NO2 relative concentration that can be detected by our SPV gassensor device with respect to the commonly used signal-to-noise (S/N) ratio, one of the mostimportant usable parameters for most commercial gas detection systems.

3.1. Influence of NO2 Gas Flow Rate at Constant Relative Concentration on Gas Sensor Response

In general, taking into account the dynamic aspects of gas adsorption/desorption effects at thesurface of gas sensor materials, the gas-sensing characteristics should strongly depend on the flowrate of target gas reaching the gas sensor system measuring chamber. This is related to the fact that

Sensors 2020, 20, 408 4 of 9

with an increased flow rate, the amount of NO2 adsorbed at the surface of the gas sensor material alsoincreases. This can be a cause of the variation of shape of response curve towards the correspondinghigher response time, as well as the shape of recovery curve towards the corresponding higherrecovery time directly related to the favorable conditions for the gas desorption. Thus, in our studywe have determined the influence of NO2 gas flow rate at constant concentration in synthetic air onthe above-mentioned gas sensor response characteristics, together with basic gas sensor dynamicparameters like response and recovery time.

Figure 2 shows the time-dependent variation of amplitude SPV signal as a function of NO2 gasflow rate in the range of 100 ÷ 20 mL/min, at the constant relative NO2 concentration of 20 ppm in thestandard dry synthetic air (with respect to the level in the synthetic air).

Sensors 2020, x  4  of  9 

the influence of NO2 gas flow rate at constant concentration in synthetic air on the above‐mentioned gas 

sensor response characteristics, together with basic gas sensor dynamic parameters like response and 

recovery time. 

Figure 2 shows the time‐dependent variation of amplitude SPV signal as a function of NO2 gas 

flow rate in the range of 10020 mL/min, at the constant relative NO2 concentration of 20 ppm in the 

standard dry synthetic air (with respect to the level in the synthetic air). 

 

Figure 2. The time‐dependent variation of amplitude of SPV signal in NO2 at various gas flow rates, 

and at a constant relative concentration of 20 ppm in synthetic air. 

One can observe from our SPV gas‐sensing characteristics that with the lowering of gas flow rate 

in the range of 10020 mL/min (at the constant NO2 relative concentration of 20 ppm in the standard 

dry synthetic air) the amplitude of SPV signal decreases only by a factor of less than 2, taking into 

account the average accuracy of determination of variation of SPV signal at the level of about a single 

mV, which  corresponds  to  the  signal‐to‐noise  (S/N)  ratio at  the  level of 3. The obtained data are 

summarized in the Table 1.   

Apart from the decreased amplitude of SPV signal, an additional effect is visible, i.e., a slight 

variation of the shape (slope) of respective gas sensor curves, directly related to the variation of gas 

sensor dynamic parameters like response and recovery time(s), respectively.   

Figure 3 presents the time‐dependent variation in the amplitude of SPV signal for the mean NO2 

gas flow rate of 60 mL/min, at the constant relative NO2 concentration of 20 ppm in the standard dry 

synthetic air, and respective gas sensing characteristics including dynamic parameters.   

As shown in Figure 3, the response and recovery time for the mean NO2 gas flow rate of 60 mL/min, 

at the constant relative concentration of 20 ppm  in the standard dry synthetic air, was ~240 s and 

~1400 s, respectively. The respective values of the above‐mentioned dynamic parameters estimated 

for the various gas flow rates are also summarized in Table 1. In general, the dynamic gas sensor 

parameters for our SPV gas sensor system at the above‐mentioned gas flow rate range of 10020 mL/min, 

look  rather poor. For  instance,  the  response  time decreased only by  about  50%. A more  evident 

tendency is observed for the recovery time because it decreased by a factor of ~3. As mentioned above, 

these effects can be correlated, from one side to the various amount of NO2 gas adsorbed at the surface 

of gas sensor materials, and from a second one, to the more favorable conditions for the desorption 

Figure 2. The time-dependent variation of amplitude of SPV signal in NO2 at various gas flow rates,and at a constant relative concentration of 20 ppm in synthetic air.

One can observe from our SPV gas-sensing characteristics that with the lowering of gas flowrate in the range of 100 ÷ 20 mL/min (at the constant NO2 relative concentration of 20 ppm in thestandard dry synthetic air) the amplitude of SPV signal decreases only by a factor of less than 2, takinginto account the average accuracy of determination of variation of SPV signal at the level of about asingle mV, which corresponds to the signal-to-noise (S/N) ratio at the level of 3. The obtained data aresummarized in the Table 1.

Table 1. Variation in the SPV signal at various NO2 gas flow rates at constant 20 ppm relativeconcentration, together with the basic dynamic parameters.

Gas Sensor CharacteristicsNO2 Gas Flow Rate (mL/min)

100 80 60 40 20

∆SPV [mV] 133 119 107 94 81

Dynamicparameters

Response time [s] ~340 ~300 ~240 ~215 ~195

Recovery time [s] ~2400 ~2000 ~1450 ~1200 ~900

Sensors 2020, 20, 408 5 of 9

Apart from the decreased amplitude of SPV signal, an additional effect is visible, i.e., a slightvariation of the shape (slope) of respective gas sensor curves, directly related to the variation of gassensor dynamic parameters like response and recovery time(s), respectively.

Figure 3 presents the time-dependent variation in the amplitude of SPV signal for the mean NO2

gas flow rate of 60 mL/min, at the constant relative NO2 concentration of 20 ppm in the standard drysynthetic air, and respective gas sensing characteristics including dynamic parameters.

Sensors 2020, x  5  of  9 

of  various  amounts  of NO2 gas  from  the  surface  of  gas  sensor materials under  the  synthetic  air 

exposure at gas flow rate 60 mL/min. 

 

Figure 3. The time‐dependent variation of SPV signal for a synthetic air flow rate of 60 mL/min and 20 ppm 

of NO2 concentration. 

Table  1. Variation  in  the  SPV  signal  at  various NO2  gas  flow  rates  at  constant  20  ppm  relative 

concentration, together with the basic dynamic parameters. 

 

Gas Sensor Characteristics   

NO2 Gas Flow Rate (mL/min) 

100    80    60    40    20 

SPV [mV]  133  119  107  94  81 

Dynamic 

parameters   

Response time 

[s] ~340  ~300  ~240  ~215  ~195 

Recovery time 

[s] ~2400  ~2000  ~1450  ~1200  ~900 

However,  it  should be noticed  that  these values have been obtained  for our SPV gas  sensor 

system at  room  temperature. Crucially,  it directly confirms  that  the  target gas  (NO2) used  in our 

experiments was only physically adsorbed onto the inner surfaces of the ZnO nanostructured thin 

films used as the gas sensor material. In such conditions a better dynamic characteristics could not 

be expected. Nevertheless, these dynamic characteristics are better than those of the commonly used 

conductometric gas sensors for which the response signal at room‐temperature working conditions 

are usually close to the signal‐to‐noise ratio. Of course, one can expect that recovery time of our gas 

sensor system will be lowered (i.e., recovery effects can be faster), but only after application of the 

additional electrode degassing mechanism (e.g., heat or light‐based). However, the main advantage 

of our SPV gas sensor system would be lost. Moreover, further study is required in order to optimize 

the degassing conditions of the target gas, which are currently in progress. 

3.2. Threshold Sensitivity of the SPV Gas Sensor System in NO2 Atmosphere 

Figure 3. The time-dependent variation of SPV signal for a synthetic air flow rate of 60 mL/min and 20ppm of NO2 concentration.

As shown in Figure 3, the response and recovery time for the mean NO2 gas flow rate of 60 mL/min,at the constant relative concentration of 20 ppm in the standard dry synthetic air, was ~240 s and ~1400 s,respectively. The respective values of the above-mentioned dynamic parameters estimated for thevarious gas flow rates are also summarized in Table 1. In general, the dynamic gas sensor parametersfor our SPV gas sensor system at the above-mentioned gas flow rate range of 100÷20 mL/min, lookrather poor. For instance, the response time decreased only by about 50%. A more evident tendencyis observed for the recovery time because it decreased by a factor of ~3. As mentioned above, theseeffects can be correlated, from one side to the various amount of NO2 gas adsorbed at the surface ofgas sensor materials, and from a second one, to the more favorable conditions for the desorption ofvarious amounts of NO2 gas from the surface of gas sensor materials under the synthetic air exposureat gas flow rate 60 mL/min.

However, it should be noticed that these values have been obtained for our SPV gas sensor systemat room temperature. Crucially, it directly confirms that the target gas (NO2) used in our experimentswas only physically adsorbed onto the inner surfaces of the ZnO nanostructured thin films used asthe gas sensor material. In such conditions a better dynamic characteristics could not be expected.Nevertheless, these dynamic characteristics are better than those of the commonly used conductometricgas sensors for which the response signal at room-temperature working conditions are usually close tothe signal-to-noise ratio. Of course, one can expect that recovery time of our gas sensor system willbe lowered (i.e., recovery effects can be faster), but only after application of the additional electrodedegassing mechanism (e.g., heat or light-based). However, the main advantage of our SPV gas sensor

Sensors 2020, 20, 408 6 of 9

system would be lost. Moreover, further study is required in order to optimize the degassing conditionsof the target gas, which are currently in progress.

3.2. Threshold Sensitivity of the SPV Gas Sensor System in NO2 Atmosphere

As our second objective we concentrated on determining the lowest NO2 relative concentration insynthetic air that can be detected by our SPV gas sensor device, with respect to the commonly usedsignal-to-noise (S/N) ratio.

Figure 4 presents the time-dependent variation in the amplitude of SPV signal for the loweredrelative concentration of NO2 target gas below 1 ppm in the standard dry synthetic air, at the meanNO2 gas flow rate of 60 mL/min.

Sensors 2020, x  6  of  9 

As our second objective we concentrated on determining the lowest NO2 relative concentration 

in synthetic air that can be detected by our SPV gas sensor device, with respect to the commonly used 

signal‐to‐noise (S/N) ratio.     

Figure 4 presents the time‐dependent variation in the amplitude of SPV signal for the lowered 

relative concentration of NO2 target gas below 1 ppm in the standard dry synthetic air, at the mean 

NO2 gas flow rate of 60 mL/min. 

 

Figure 4. The time‐dependent variation in the SPV signal for lowered NO2 relative concentration in 

synthetic air at mean constant gas flow rate 60 mL/min. 

As can be seen from Figure 4, the decreased relative concentration of NO2 is accompanied by a 

decreased relative amplitude in the SPV signal, reaching its lowest value for of NO2 at the level of 125 ppb, 

taking into account the commonly used criteria that the level of signal‐to‐noise (S/N) ratio should not 

be lower than 3. Crucially, this value was already reached at room temperature, which looks very 

promising with  respect  to  the  performance  of  commonly  used  conductometric  type  gas  sensor 

systems.  For  instance,  similar  experiments  have  been  performed  by  Procek  et  al.  [28]  for  the 

conductometric  type gas  sensor device using  similar ZnO nanostructures and  the additional UV 

radiation of similar power density, but working mainly at a higher temperature (~200 C). However, 

there is only one experimental point on the gas sensor response to 1 ppm of NO2 at room temperature, 

that can be compared with our results. The obtained response was 304  74 (%), which corresponds 

to similar signal‐to‐noise (S/N) ratio  4, as used in our analysis. In addition to the above, it should be noted that, from our SPV gas‐sensing characteristics shown 

in Figure 4, one also notices that the shape of the response curves is similar. A more precise analysis 

of their sensor dynamic characteristics proved that the gas response time(s) are similar at the level of 

about 1000 s, respectively, which was longer with respect to the higher relative concentration of NO2 

in the synthetic air.   

Of additional importance, the respective recovery time(s) are similar at the level of about 1000 s, 

and comparable with the gas response time(s). It corresponds to the respective gas sensor dynamic 

parameters  for  the  gas  response  curves  at  lower  gas  flow  rates  of NO2  at  the  constant  relative 

concentration in the synthetic air shown in Figure 2. However, the tendency for gas sensor dynamic 

Figure 4. The time-dependent variation in the SPV signal for lowered NO2 relative concentration insynthetic air at mean constant gas flow rate 60 mL/min.

As can be seen from Figure 4, the decreased relative concentration of NO2 is accompanied by adecreased relative amplitude in the SPV signal, reaching its lowest value for of NO2 at the level of 125ppb, taking into account the commonly used criteria that the level of signal-to-noise (S/N) ratio shouldnot be lower than 3. Crucially, this value was already reached at room temperature, which looks verypromising with respect to the performance of commonly used conductometric type gas sensor systems.For instance, similar experiments have been performed by Procek et al. [28] for the conductometrictype gas sensor device using similar ZnO nanostructures and the additional UV radiation of similarpower density, but working mainly at a higher temperature (~200 C). However, there is only oneexperimental point on the gas sensor response to 1 ppm of NO2 at room temperature, that can becompared with our results. The obtained response was 304 ± 74 (%), which corresponds to similarsignal-to-noise (S/N) ratio ≈ 4, as used in our analysis.

In addition to the above, it should be noted that, from our SPV gas-sensing characteristics shownin Figure 4, one also notices that the shape of the response curves is similar. A more precise analysis oftheir sensor dynamic characteristics proved that the gas response time(s) are similar at the level of

Sensors 2020, 20, 408 7 of 9

about 1000 s, respectively, which was longer with respect to the higher relative concentration of NO2 inthe synthetic air.

Of additional importance, the respective recovery time(s) are similar at the level of about 1000 s,and comparable with the gas response time(s). It corresponds to the respective gas sensor dynamicparameters for the gas response curves at lower gas flow rates of NO2 at the constant relativeconcentration in the synthetic air shown in Figure 2. However, the tendency for gas sensor dynamicparameters to be directly related to the controlled adsorption of gas species at the surface of gas sensormaterial is described above.

4. Conclusions

In this study the working conditions of our novel, high-sensitivity gas sensor device using thesurface photovoltage effect, based on the Kelvin probe approach, have been elaborated.

We have focused on the determination of the influence of gas flow rate of NO2 at the constantrelative concentration of 20 ppm on the sensor response and recovery times, respectively, as well as onthe determination of lowest relative NO2 concentration below 1 ppm relative concentration in syntheticair that can be detected with this sensor.

Firstly, it was observed that with the lowering of target gas flow rate in the range of 100÷20 mL/min,the amplitude of SPV signal decreases only by a factor of less than 2, taking into account the averageaccuracy of determination of variation of the SPV signal at the level of about single mV. Moreover, aslight variation in the shape of respective gas sensor curves was observed directly related to a variationof gas sensor dynamic parameters, and almost twice decreasing of the respective response/recoverytime(s) was determined.

Secondly, it has been established that, with the lowering of relative concentration of NO2 in thestandard dry synthetic air, an evident tendency towards decreasing of the relative SPV signal appears,reaching at room temperature a smallest relative NO2 concentration at the level of 125 ppb, with respectto signal-to-noise (S/N) ratio at the level of 3. Moreover, the gas sensor response time(s) are similar tothe respective recovery time(s) being at the level of about 1000 s.

This study confirms that our SPV gas sensor device [19] has the important advantages thatmake it potentially suitable for wide practical application. In an ongoing study we plan to improvethe performance of our SPV gas detector system by using other specific MOX low dimensionalnanostructures recently elaborated by our group [29–31] with extended internal surfaces. We anticipatethat, with the more-effective adsorption/desorption effects of target gases, one can expect to attainbetter responses as well as reduction of the response and recovery times. Moreover, we also plan toelaborate the effective source for the faster removal of specific target gas species from the surface of gassensor material(s) during the regeneration process in our of SPV gas detector system.

Author Contributions: Conceptualization, M.K. and J.S.; methodology, M.K. and J.S.; investigation, M.K. and J.S.;writing–original draft preparation, M.K.; writing-review & editing, M.K. and J.S..; project administration, M.K.;funding acquisition, M.K. All authors have read and agreed to the published version of the manuscript.

Funding: This work was realized within the Statutory Funding of the Department of Cybernetics, Nanotechnologyand Data Processing, Silesian University of Technology, Gliwice, Poland, financed by the research grant of NationalScience Centre, Poland—OPUS 11, No. 2016/21/B/ST7/02244, and additionally M.K. has been supported from theProfessor Grant (GP) of the Silesian University of Technology no. 02l0504GP19/0050.

Conflicts of Interest: The authors declare no conflict of interest.

References

1. Comini, E.; Faglia, G.; Sberveglieri, G. Electrical-based gas sensing. In Solid State Gas Sensing; Springer:Boston, NY, USA, 2009; ISBN 978-0-387-09664-3.

2. Liu, X.; Cheng, S.; Liu, H.; Hu, S.; Zhang, D.; Ning, H. A survey on gas sensing technology. Sensors 2012, 12,9635–9665. [CrossRef]

Sensors 2020, 20, 408 8 of 9

3. Korotcenkov, G. Metal oxides for solid-state gas sensors: What determines our choice? Mater. Sci. Eng. B2007, 139, 1–23. [CrossRef]

4. Comini, E.; Baratto, C.; Faglia, G.; Ferroni, M.; Vomiero, A.; Sberveglieri, G. Quasi one dimensional metaloxide semiconductors: Preparation, characterization and application as chemical sensors. Progr. Mater. Sci.2009, 54, 1–67. [CrossRef]

5. Fine, G.F.; Cavanagh, L.M.; Afonja, A.; Binions, R. Metal Oxide Semi-Conductor Gas Sensors in EnvironmentalMonitoring. Sensors 2010, 10, 5469–5502. [CrossRef]

6. Baratto, C.; Comini, E.; Faglia, G.; Sberveglieri, G. The Power of Nanomaterial Approaches in Gas Sensors.In Solid State Gas Sensors—Industrial Application; Fleischer, M., Lehmann, M., Eds.; Springer: Berlin-Heidelberg,Germany, 2011; pp. 53–78.

7. Carpenter, M.A.; Mathur, S.; Kolmakov, A. Metal Oxide Nanomaterials for Chemical Sensors; Springer: Boston,NY, USA, 2012; ISBN1 146145395X. ISBN2 9781461453956.

8. Eranna, G. Metal Oxide Nanostructures as Gas Sensing Devices; CRC Press: Boca Raton, FL, USA, 2012; ISBN978-1-4398-6340-4.

9. Korotcenkov, G.S. Handbook of Gas Sensor Materials: Properties, Advantages and Shortcomings forApplications. In New Trends and Technologies; Springer: Boston, NY, USA, 2013; ISBN 978-1-4614-7387-9.

10. Dey, A. Semiconductor metal oxide gas sensors: A review. Mater. Sci. Eng. B 2018, 229, 206–217. [CrossRef]11. Zhu, L.; Zeng, W. Room-temperature gas sensing of ZnO-based gas sensor: A review. Sens. Actuators A 2017,

267, 242–261. [CrossRef]12. Comini, E. Metal oxide nanowire chemical sensors: Innovation and quality of life. Mater. Today 2016, 19,

559–567. [CrossRef]13. Mizsei, J. Vibrating capacitor method in the development of semiconductor gas sensors. Thin Solid Films

2005, 490, 17–21. [CrossRef]14. Oprea, A.; Barsan, N.; Weimar, U. Work function changes in gas sensitive materials: Fundamentals and

applications. Sens. Actuators B 2009, 142, 470–493. [CrossRef]15. Korotcenkov, G.; Cho, B.K. Porous Semiconductors: Advanced Material for Gas Sensor Applications. Crit. Rev.

Solid State Mater. Sci. 2010, 35, 1–37. [CrossRef]16. Kronik, L.; Shapira, Y. Surface photovoltage phenomena: Theory, experiments, and applications. Surf. Sci. Rep.

1999, 37, 1–206. [CrossRef]17. Rothschild, A.; Levakov, A.; Shapira, Y.; Ashkenasy, N.; Komem, Y. Surface photovoltage spectroscopy study

of reduced and oxidized nanocrystalline TiO2 films. Surf. Sci. 2003, 456–460. [CrossRef]18. Sivalingam, Y.; Magna, G.; Pomarico, G.; Martinelli, E.; Paolesse, R.; D’Amico, A.; Di Natale, C. Gas effect on the

surface photovoltage of porphyrins functionalized ZnO nanorods. Adv. Mater. Lett. 2012, 3, 442–448. [CrossRef]19. Kwoka, M.; Borysiewicz, M.A.; Tomkiewicz, P.; Piotrowska, A.; Szuber, J. A novel type room temperature

surface photovoltage gas sensor device. Sensors 2018, 18, 2919. [CrossRef] [PubMed]20. Eklöv, T.; Lundström, I. Gas mixture analysis using a distributed chemical sensor system. Sens. Actuators

B Chem. 1999, 57, 274–282. [CrossRef]21. Lezzi, A.M.; Beretta, G.P.; Comini, E.; Faglia, G.; Galli, B.; Sberveglieri, G. Influence of gaseous species

transport on the response of solid state gas sensors within enclosures. Sens. Actuators B 2001, 78, 144–150.[CrossRef]

22. Utriainen, M.; Kärpänoja, E.; Paakkanen, H. Combining miniaturized ion mobility spectrometer and metaloxide gas sensor for the fast detection of toxic chemical vapors. Sens. Actuators B 2003, 93, 17–24. [CrossRef]

23. Frank, K.; Kohler, H.; Guth, U. Influence of the measurement conditions on the sensitivity of SnO2 gassensors operated thermo-cyclically. Sens. Actuators B 2009, 141, 361–369. [CrossRef]

24. Gmür, R.; Goschnick, J.; Hocker, T.; Schwarzenbach, H.; Sommer, M. Impact of sensor packaging on analyticalperformance and power consumption of metal oxide based gas sensor microarrays. Sens. Actuators B 2007,127, 107–111. [CrossRef]

25. Righettoni, M.; Tricoli, A.; Gass, S.; Schmid, A.; Amann, A.; Pratsinis, S.E. Breath acetone monitoring byportable Si: WO3 gas sensors. Anal. Chim. Acta 2012, 738, 69–75. [CrossRef]

26. Borysiewicz, M.A.; Dynowska, E.; Kolkovsky, V.; Dyczewski, J.; Wielgus, M.; Kaminska, E.; Piotrowska, A.From porous to dense thin ZnO films through reactive DC sputter deposition onto Si (100) substrates.Phys. Status Solidi A 2012, 209, 2463–2469. [CrossRef]

Sensors 2020, 20, 408 9 of 9

27. Masłyk, M.; Borysiewicz, M.A.; Wzorek, M.; Wojciechowski, T.; Kwoka, M.; Kaminska, E. Influence ofabsolute argon and oxygen flow values at a constant ratio on the growth of Zn/ZnO nanostructures obtainedby DC reactive magnetron sputtering. Appl. Surf. Sci. 2016, 389, 287–293. [CrossRef]

28. Procek, M.; Stolarczyk, A.; Pustelny, T. Impact of temperature and UV irradiation on dynamics of NO2

sensors based on ZnO nanostructures. Nanomaterials 2017, 7, 312. [CrossRef] [PubMed]29. Kwoka, M.; Krzywiecki, M. Rheotaxial growth and vacuum oxidation—Novel technique of tin oxide

deposition—In situ monitoring of oxidation process. Mater. Lett. 2015, 154, 1–4. [CrossRef]30. Kwoka, M.; Galstyan, V.; Comini, E.; Szuber, J. Pure and highly Nb-doped titanium dioxide nanotubular

arrays: Characterization of local surface properties. Nanomaterials 2017, 7, 456. [CrossRef] [PubMed]31. Kwoka, M.; Lyson-Sypien, B.; Kulis, A.; Maslyk, M.; Borysiewicz, M.A.; Kaminska, L.; Szuber, J. Surface

properties of nanostructured porous ZnO thin films prepared by direct current reactive magnetron sputtering.Materials 2018, 11, 131. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (http://creativecommons.org/licenses/by/4.0/).